[sdiy] VCA troubles, NE5517

[ASSI]

On Wednesday 14 April 2004 16:56, Fredrik Carlqvist wrote:
> Vdd is 5 volts. Vdd/2 is 2.5V. I didn't use the buffer since I
> am using a single +5V supply, and I didn't want the 1.2V drop over
> the darlington.

You need 3.5V just for biasing the NE5517 properly, that does not leave 
much room for your signal. More importantly, by fixing one input to 
Vdd/2, you use up 2.1V just to the ground. At these biasing conditions 
the difference in speed between npn and pnp mirrors may show and 
produce the feedthrough your experiencing. You could try produce the 
output voltage and positive input bias with 2.2k resistors to Vdd and 
Gnd and bias the negative input with two 200k resistors also from Vdd 
and Gnd to balance the input impedances. But since you're not using the 
extra features and the other half of the NE5517 anyway, popping in a 
CA3080 might be worth a thought.


Achim.
-- 
+<[Q+ Matrix-12 WAVE#46 microQkb/Omega Andromeda XTk/30 sonic heaven]>+

SD adaptations for KORG EX-800 and Poly-800MkII V0.9:
http://homepages.compuserve.de/Stromeko#KorgSDada